Skip links

Design Considerations in Nonclinical Development of Cell and Gene Therapies

Designing an adequate nonclinical program to support the safety of cell or gene therapy products is not always straightforward. In this blog, we will address common questions we receive from our clients and aspects to consider when designing those nonclinical studies.

In which species should the toxicity study be conducted? Does it have to be in a disease model?

The cell or gene therapy should be pharmacologically active in the animal model. If the target is only active in a disease state, then efficacy and safety studies may need to be conducted in the disease model. If this is the case, consider collecting both efficacy and safety endpoints in the same study to reduce the use of animals. If the compound is not active in any species (eg, human-specific gene target), a surrogate molecule could be used. Note that there is no strict requirement of toxicity studies in two species, which is often needed for small molecule programs.

Do you have a unique or novel route of administration (RoA) (eg, surgical procedure, intracerebral)? If so, consider using an animal species that will accommodate that RoA or provide a way to best mimic that RoA.

In the end, the data generated needs to provide biologically relevant information about how the therapy will behave in humans. Provide justifications for the species used and understand the limitations of each model and how data from animal studies translate to humans. These will be important to outline clearly in the Investigational New Drug application (IND) or Clinical Trial Application (CTA).

 How long does the study need to be?

Acute and delayed toxicity should be evaluated in safety studies. Thus, it is important to know the kinetics of your therapy (eg, how long does it stay in the body, how long is it active, when is peak expression). Then, build in timepoints into the safety study to provide meaningful data on potential toxicity. For example, evaluate acute toxicity at the peak of activity, delayed toxicity at a timepoint where there is steady state of expression/activity, and, if needed, recovery at a timepoint that provides enough time for recovery, taking into consideration the rate of clearance of therapeutic or cessation of activity.

 What endpoints need to be evaluated?

Toxicity studies should strive to evaluate all typical general toxicity endpoints (eg, clinical observations, clinical pathology, histology). In addition, immunogenicity endpoints (eg, anti-drug antibodies, cytokine production, T-cell response to vector and transgene) are also typically needed to understand the body’s response to the therapeutic. Biodistribution of the product (transgene/gene or cells) should be evaluated. Remember to consider if clinical data on similar therapies can help inform target organs or reactions to monitor for in your studies.

Consider including vector shedding (eg, in feces or urine); though there is a discrepancy between some regulatory agencies, it seems that the field is moving in the direction of expecting vector shedding data, so it is best to collect the samples just in case.

The need for safety pharmacology, genotoxicity (in the form of evaluating potential for integration), reproductive toxicity, and tumorgenicity should be assessed based on the target indication and nature of the therapy. For example, if there is no biodistribution to reproductive organs, then conducting reproductive toxicity studies may not be needed.

 What doses should I use for safety studies?

The nonclinical program for cell or gene therapy products should identify a potentially efficacious dose and ideal dose regimen for the proposed indication while also characterizing potential toxicity profiles and target organs. Doses evaluated in the safety study should bracket the expected clinical dose level range. For example, the low dose is the minimally effective dose, the mid dose is the highest anticipated clinical administered dose, and the high dose is some multiple of the mid dose to provide some margin of safety.

Note that first-in-human doses should be selected using all available information, including clinical data. For example, CAR-T cell products often use clinical data from similar products to inform starting doses.


The nonclinical program to support safety of cell and gene therapy products should be tailored to each specific therapy. It is important to understand the biology of your therapeutic in order to design an adequate safety assessment. Talk early and often with regulatory agencies to gain agreement on the design and breadth of your proposed nonclinical safety studies; these interactions de-risk the program, provide you with the most up-to-date information on regulatory expectations for your therapeutic, and ensure patient safety.

Still not sure? Toxicologists at Premier Consulting can help you design and/or conduct appropriate nonclinical studies to support your cell or gene therapy product. Contact us today.


Author: Madelyn Huang, PhD, DABT, Toxicologist